Rio de Janeiro's "homegrown" LLM appears to be a merge of an existing model

Rio de Janeiro’s “homegrown” LLM appears to be a merge of an existing model

里约热内卢的“自研”大模型被指实为现有模型的合并产物

nex-agi / Nex-N2 Public Notifications You must be signed in to change notification settings Fork 12 Star 177 Rio-3.5-Open-397B ≈ 0.6 x Nex-N2_pro + 0.4 x Qwen #4New issueCopy linkNew issueCopy linkOpenOpenRio-3.5-Open-397B ≈ 0.6 x Nex-N2_pro + 0.4 x Qwen#4Copy linkDescription00INDEXopened on Jun 14, 2026Issue body actions (注：此部分为 GitHub 仓库元数据，已忽略无关导航文本)

prefeitura-rio/Rio-3.5-Open-397B is presented as an original 397B model trained by IplanRIO. It is not. Its weights are a direct element-wise merge of our model, Nex, with the official Qwen3.5-397B-A17B base — about 0.6 Nex / 0.4 Qwen — and we find no evidence of any training of their own. prefeitura-rio/Rio-3.5-Open-397B 被宣传为由 IplanRIO 训练的原创 397B 参数模型。事实并非如此。其权重实际上是我们模型 Nex 与官方 Qwen3.5-397B-A17B 基座的直接逐元素合并（比例约为 0.6 Nex / 0.4 Qwen），我们没有发现任何他们进行过自主训练的证据。

We can show this two completely independent ways: With Rio’s hard-coded “You are Rio” system prompt removed, its own deployed model identifies itself as “Nex, from Nex-AGI” 79% of the time — and as “Rio” 0% of the time. It even recites our organization’s bespoke backstory word-for-word. 我们可以通过两种完全独立的方式证明这一点：在移除 Rio 硬编码的“你是 Rio”系统提示词后，其部署的模型在 79% 的情况下会将自己标识为“来自 Nex-AGI 的 Nex”，而标识为“Rio”的概率为 0%。它甚至能逐字背诵我们组织定制的背景故事。

Every weight tensor in Rio is, to thousands of standard deviations, the same 0.6/0.4 blend of Nex and Qwen — across all 60 layers and every component of the network. Other finetunes cannot be explained as interpolations. Below is the evidence. Judge for yourself. Rio 中的每一个权重张量，在数千个标准差的范围内，都与 Nex 和 Qwen 的 0.6/0.4 混合比例完全一致——这涵盖了所有 60 层网络及其每一个组件。其他的微调模型无法用这种插值法来解释。以下是证据，请自行判断。